complement activation - nanomedicinelabofnano.gmu.edu/.../book-chapter-on-complement-system.pdf ·...

357

b1429 Handbook of Immunological Properties of Engineered Nanomaterials

Chapter 11

Complement Activation

Carolina Salvador-Morales*,†,§ and Robert B. Sim‡

*Bioengineering Department, George Mason University

4400 University Drive, MS 1G5 Fairfax

VA 22030, USA†Krasnow Institute for Advanced Study

George Mason University, 4400 University Drive

MS 2A1 Fairfax, VA 22030, USA‡Department of Pharmacology, University of Oxford

Mansfield Road, OX1 3QT, UK

The complement system is the most important biochemical cascade in the blood for the

recognition, opsonization, and elimination of foreign materials. To date, the leading

causes of death in the United States include cancer, cardiovascular and neurodegenerative

diseases, and diabetes. New treatments are urgently needed to treat these devastating

diseases and nanotechnology potentially provides new avenues to fi ght such illnesses.

These avenues include the development of novel nanocarriers that deliver drugs in a

specifi c and controlled manner, while minimizing secondary effects. The success of

bioengineering effective nanocarriers for drug delivery purposes requires a deep

understanding of the interaction between the complement system and the nanocarriers.

This review focuses on reporting the current state of complement activation by different

nanomaterials. Here, we assess various important parameters that infl uence the activation

of the complement system, which include the physicochemical characteristics of both

nanocarriers and complement proteins. We next evaluate the most recent engineering

approaches to prevent or reduce complement activation. Finally, we discuss different

in vitro and in vivo procedures to assess complement activation.

§Corresponding author. E-mail: [email protected]

358 C. Salvador-Morales & R. B. Sim


1. The Complement System

The complement system is a group of about 35 soluble and cell surface

proteins in blood and other body fluids which interact to recognize, opsonize,

and clear or kill invading microorganisms, altered host cells, and other foreign

materials, including many synthetic materials.1 A simplified representation of

the complement system is shown in Figure 1. The activation of the comple-

ment system can occur by any of three pathways, termed the classical, lectin,

and alternative pathways.

1.1. Pathways of complement activation

1.1.1. The classical pathway

In the classical pathway (Figure 1), the recognition protein C1q binds to

targets (activators), and the binding causes two proteases, C1r and C1s,

attached to C1q, to become active. When C1s is activated, it cleaves and

activates the next two proteins of the system, which are called C4 and C2.

Domains of these proteins then form a complex called C4b2a, which is itself

a protease that cleaves and activates the most abundant complement protein in

blood, C3. C3 is activated to form C3b, which binds back on to the surface

Figure 1. The pathways of the complement system. The complement system can be acti-

vated via three pathways, namely the classical, lectin, and alternative pathways.

Complement Activation 359


of the target. The target becomes coated with clusters of hundreds of C3b

molecules, which gradually are cleaved by other proteases to forms of C3

called iC3b or C3d/C3dg.

C3b is recognised by CR1 (complement receptor 1) which is on red

blood cells and some white blood cells, and this causes the target to bind

to the cells. A C3b-coated target, bound to red blood cells, can circulate

in the blood, with the C3b gradually being converted to iC3b. As the red

blood cell and target pass through the liver or spleen, they come into

contact with macrophages, which have on their surface receptors for iC3b

(named CR3 and CR4). The target, with its attached iC3b, is stripped off

of the red blood cell and transferred to the macrophage, which ingests the

target and destroys it intracellularly. This phenomenon, by which targets

are coated with C3b or iC3b that promote their uptake and destruction

by macrophages, is called opsonization (Figure 1). When iC3b is further

broken down to C3d/C3dg, it interacts with another receptor, CR2,

which is present on B lymphocytes. This interaction can stimulate the

synthesis of antibodies against the target. Dendritic cells, which capture

foreign materials and present them to the adaptive immune system as anti-

gens, also have complement receptors, thus coating of the target with C3

fragments also helps to develop the adaptive immune response (antibodies

and cytotoxic T cells) against the target. These activities that promote the

antibody or T cell response are called the adjuvant activities of comple-

ment (Figure 1).

Once C3 has been activated, the protease C4b2a can then activate the

next complement protein C5, forming the fragments C5a and C5b. C5b

binds to C6, C7, C8, and C9, and this large protein complex called the MAC

(membrane attack complex) can insert itself into lipid bilayers (cell mem-

branes). If the target has a cell membrane, the MAC will effectively make

holes in the membrane, killing the target (“lysis”) (Figure 1).

During the activation of C4, C3, and C5, small peptides C4a, C3a, and

C5a, collectively called anaphylotoxins, are released, and these potentially

have inflammatory effects. They affect the smooth muscle of blood vessels

and cause fluid leakage from the blood into the tissue spaces. They have

effects on cytokine and chemokine release, and C5a is also a chemotactic fac-

tor, i.e., it attracts cells (i.e., granulocytes). At the site of a wound, these

activities would cause the leakage of fluid into the wound, releasing more

complement proteins from the blood into the site, to opsonize infectious

microorganisms and also cause granulocytes to migrate to the site, where they

ingest and kill bacteria.



1.1.2. The lectin pathway

The lectin pathway is initiated by the binding of mannan-binding lectin

(MBL) or ficolins to carbohydrate structures present on a wide range of

microorganisms including bacteria, viruses, fungi, and parasites. MBL has

also been reported to bind to IgA and to agalactosyl IgG (IgG-Go). MBL

and ficolins circulate in serum complexed with serine protease proenzymes

called MBL-associated serine proteases (MASPs) and a small 19-kDa related

protein (Map 19). MASPs are structurally similar to C1r and C1s with identi-

cal domain composition. The mechanisms for MASP activation have yet to be

fully determined but upon MBL of ficolin binding to targets, MASP-2 is

auto-activated and cleaves C4 and C2 to form the C3 convertase C4b2a,

similar to that in the classical pathway.

1.1.3. The alternative pathway

The alternative complement pathway is initiated differently from the classical

and lectin pathways. To trigger this pathway, C3b has to be deposited on the

surface of a target. C3b may be derived from the activation of the other path-

ways, arise from a non-enzymic slow turnover of C3, or it may arise because

other non-complement proteases can activate C3 to a minor extent. Once

one molecule of C3b has bound to the target surface, factor B can bind to it,

and is then cleaved by factor D to form C3bBb, which is a protease complex

homologous to C4b2a in the other pathways. Once this C3-cleaving enzyme

has formed, events occur as per the classical and lectin pathways. Since C3bBb

can form more molecules of C3b, the alternative pathway acts as an amplifica-

tion loop (Figure 1), causing more C3b to be produced and deposited on the

target surface.

The different pathways of complement respond to different targets. C1q

in the classical pathway binds to a very wide range of targets. These include

antibody–antigen complexes formed with IgG or IgM, Gram-negative bacte-

ria, some viruses, polyanionic molecules like DNA/RNA, or anionic phos-

pholipid micelles, altered host proteins such as clots or amyloids, and many

synthetic materials such as carbon nanotubes (Perspex). C1q recognises

mainly charged clusters, but probably also hydrophobic patches on surfaces.

1.2. Physicochemical characteristics of key complement proteins

Among 3,700 different proteins in human serum,2 only a handful play a role

in the activation of the complement system. C1q is the recognition molecule



for the classical pathway, MBL or the ficolins act for the lectin pathway, while

C3 and C3b are key molecules for the alternative pathway. The direct or indi-

rect binding of C1q3 and C3b3 to a carbon nanotube surface activates the

complement system. The extent of the complement activation induced by a

nanomaterial will depend on the physicochemical characteristics of the nano-

material and its interaction with complement proteins. Here, we discuss the

most important physicochemical characteristics of C1q, C3, and C3b that

activate the complement system when interacting with foreign materials.

1.2.1. Morphology of C1q

C1q is a 460-kDa protein composed of six heterotrimeric collagen-like triple

helices that converge in their N-terminal half to form a stalk, then diverge

to form individual stems, each terminating in a C-terminal heterotrimeric

globular domain4 (Figure 2).

C1q binds to target ligands via the globular domains, or heads, triggering

the activation of C1r and C1s, the proteases associated with C1q.5 One of the

key features of C1q morphology in initiating the complement system via the

classical pathway is that its hexameric structure allows binding by multiple

heads. Each head has three lobes, made up of homologous domains (called

A, B, and C lobes). Each has distinct but overlapping binding specificity. C1q

binds to targets by multiple weak binding interactions. Recognition of one

motif (a charge cluster or hydrophobic patch) by one lobe of a head is a weak

interaction, but since there are six heads and 18 lobes, multiple interactions

Figure 2. The structure of C1q. C1q is made up of three types of polypeptide chains — A,

B, and C — which are homologous to each other. One of each type of chain interact with one

another and intertwine to form a collagen triple-helical stalk and a three-lobed globular head.

Six of these subunits assemble to form the umbel shape shown here. MBL and the ficolins have

similar subunit structures, and form final assemblies with three–six subunits (i.e., three–six

heads).



can form, and this results in high avidity binding. A requirement for high

avidity binding is that the motifs recognized are in a certain form of regularly

spaced array on the target surface. The spacing between motifs must corre-

spond to the spacing between lobes (about 5 nm) or between heads (variable

because of the flexibility of the whole molecule) of about 10–40 nm. C1q

globular heads are approximately 6 nm wide and have multivalent charged

groups for target recognition.6 The whole C1q molecule measures approxi-

mately 40 nm in diameter. C1q morphology and its dimensions are relevant

features in understanding its interaction with invaders and foreign materials.

We have earlier reported the direct binding of C1q to double-walled carbon

nanotubes (DWNTs).3 C1q binds to DWNTs presumably through its globu-

lar heads, activating the complement system via the classical pathway. High-

pressure monoxide single-walled carbon nanotubes (HIPco SWNTs) and

multi-walled carbon nanotubes (MWNTs) also activate the complement sys-

tem via the classical pathway. Although the direct binding of C1q to HIPco

SWNTs and MWNTs has not been shown, it is likely that C1q binds to them

since these carbon nanotubes are not chemically modified, and therefore have

a large hydrophobic surface area. Ling et al.7 in a series of transmission elec-

tron microscope studies, reported the binding of C1q and C1s–C1r–C1r–

C1s to MWNTs, although no binding of C1q to either DWNTs or SWNTs

was observed in their study. They observed binding of C1s–C1r–C1r–C1s to

DWNTs, but not to SWNTs. In summary, Ling’s team did not observe bind-

ing of C1q to SWNTs and DWNTs in contrast to our previous studies.3

Although these two separate findings disagree with each other, we provide

strong evidence of the binding of C1q to DWNTs.3 For example, we meas-

ured the direct binding of C1q to DWNTs in the presence and absence of

human serum proteins. C1q binds to DWNTs in both conditions. This result

addressed the argument given by Ling and co-workers to explain the discrep-

ancies in the findings. Ling’s team believes that the SWNT and DWNT stud-

ies reported by us activated the complement system via the classical pathway

because of the formation of a serum protein layer on DWNTs. We, however,

clearly showed that C1q binds to DWNTs even in the absence of serum pro-

teins. We believe that these differences are mainly due to the surface chemis-

try of these carbon nanotubes. A careful characterization of the SWNTs and

DWNTs used in Ling’s study would be extremely useful to further explain the

disparate results.

There is a vacuum of knowledge with respect to the direct binding of C1

to core-shell nanoparticles — the preferred choice of nanoparticle design for

drug delivery purposes. These particles are often coated with a layer of poly-

ethylene glycol (PEG), which is an uncharged hydrophilic polymer that helps



reduce protein adsorption. The dimensions and the morphology of C1q

become relevant for its interaction with core-shell nanoparticles since the

whole C1q molecule is about 40 nm wide (Figure 2). Core-shell nanoparti-

cles are mainly covered with PEG units in which the distance between each

unit can be around 5 nm. Polymeric nanoparticles synthesized with

poly(lactic-co-glycolic acid) (PLGA) and polylactic acid (PLA) are usually

coated with PEG. Given these dimensions, it will be difficult for the C1 com-

plex or C1q to reach the core even when PEG units are flexible. Nevertheless,

if the surface of the core-shell particles is partially covered with PEG, C1 and

C1q will most likely interact with the core via hydrophobic interactions and

will therefore initiate the complement system activation via the classical path-

way. C3 is another protein of the complement system that plays a key role in

the complement activation cascade and will be discussed in detail in the next

section.

1.2.2. Morphology and chemistry of C3 and C4

C3 is the central and most abundant protein of the complement system,8 and

it is present in the plasma at a concentration of 1.3 g/L.8 It is a protein of

185 kDa with dimensions 15.2 nm × 9 nm × 8.4 nm9 and contains an internal

thioester bond. When C3 is activated to C3b, the thioester is exposed and is

highly reactive, and can bind covalently to surfaces (e.g., the target) bearing

NH2 and OH groups, forming amide and ester bonds, respectively. It is gen-

erally agreed that this covalent binding reaction is the main route by which

C3b becomes attached to biological targets. The same mechanism applies to

C4b, which is a homolog of C3b.10 For synthetic materials, however, there

may not be suitable surface groups to which C3b and C4b can form covalent

bonds. It is possible that other plasma proteins which bind to synthetic mate-

rials would provide such chemical groups. However, for carbon nanotubes,

C3 and C4 fragments do bind to them, but we found no evidence that they

were covalently bound to other proteins; instead, it appears that C3b and

C4b bind by hydrophobic, non-covalent interactions.11

Recent findings report that although C3 is a large protein, it is able to

interact with the diffuse shell of core-diffuse shell-structured nanoparticles.12

In this case, complement activation takes place by the complex formed

between C3 and bovine serum albumin (BSA), where the latter is able to

reach the core of the particle. Other studies report that if C3b binds to or

becomes trapped between surface PEG “brushes,” C3 hydration and confor-

mational changes (C3 tickover) may become accelerated, leading to the

assembly of fluid phase C3 convertase.13



The assessment of the complement system via the alternative pathway

might be more relevant from a clinical point of view than the assessment of

the classical pathway since there are more reports which discuss the initiation

of the alternative pathway via the binding of C3b and C3 convertase on the

surface of artificial materials.14–16 Surface plasmon resonance (SPR) is a quan-

titative technique that can measure the amount of C3b bound to the nano-

materials as reported by Toda et al.17

2. Physicochemical Characteristics of Nanoparticles

The physicochemical characteristics of different nanocarriers influence the

adsorption of human serum and plasma proteins which in turn, might include

those which activate the complement system cascade. The coating of different

nanomaterials such as liposomes, carbon nanotubes, and polymeric nanopar-

ticles with PEG is very well known as an effective way to prevent protein

adsorption. The success of this technical strategy depends on the density of

the coating (e.g., maximizing the surface coverage of the core), the length,

and the configuration of the PEG (e.g., mushroom, brush, or mushroom–

brush). These parameters will be examined carefully in the next section.

2.1. PEG chain density

As mentioned above, the decoration of a particle’s surface by covalently graft-

ing, entrapping, or adsorbing PEG chains diminishes protein adsorption.

Each of these methods has its advantages and disadvantages. For instance,

one of the biggest advantages of the covalent binding of the PEG chain to

the nanoparticle surface is that it prolongs the nanoparticle’s circulation half-

life. However, the biggest drawback of this method is that it does not ensure

complete surface coverage. In contrast, the adsorption method overcomes

this disadvantage but poses new challenges which include the desorption of

PEG chains from the nanoparticle’s surface. The desorption of PEG will

eventually cause the precipitation of the nanoparticles, preventing further use

of these particles.

The spatial distribution of PEG on a nanoparticle surface is also relevant

to protein adsorption and consequently important to diminishing or pre-

venting complement activation. Research has demonstrated that PEG chains

can take on two main different spatial arrangements on a nanoparticle sur-

face, referred to as the mushroom and brush configurations.18 In the mush-

room configuration, particles present very low surface coverage of PEG

chains. On the contrary, particles holding a brush configuration have a high



surface coverage of PEG chains. The prevention or diminution of protein

adsorption will not only depend on balancing the PEG chain density on the

particle surface, but also on other factors. For instance, high PEG chain

density ensures the complete coverage of the nanoparticle surface, but

decreases the mobility of the PEG chains, thus diminishing the steric hin-

drance properties of the PEG layer.19 To date, there is no rule of thumb for

obtaining this delicate balance. Some scientists believe that the optimal

thickness of hydrodynamic layer effectively shielding the particle surface

from protein adsorption is 5% of the particle’s hydrodynamic size. According

to another view, the optimal thickness of hydrodynamic layer should be at

least twofold greater than the radius of the polymer coil at the polymer con-

formation in its diluted solution.19

The physicochemical characteristics of the proteins are factors that

also influence protein adsorption. New studies report that the shell of core-

diffuse shell structured nanoparticles synthesized with PIBCA [poly

(isobutylcyanoacrylate)]–dextran block copolymers by a self-assembly pro-

cess12 do not prevent the adsorption of small flexible proteins such as BSA,

even when the density of chains within the diffuse shell is quite high. PIBCA

is a bioerodible and bioeliminable polymer. Large proteins such as fibrino-

gen (340 kDa) and C3 (185 kDa) interact with these nanoparticles in a

specific manner. Fibrinogen not only reaches the diffuse shell of the

nanoparticles, but also reaches their hydrophobic core. On the other hand,

C3b penetrates the diffuse shell, but does not reach the core of the

nanoparticle.20 The authors observed complement activation on nanoparti-

cles that trapped C3b in their diffuse shell. They hypothesized that

complement activation in this case is due to a complex formed between

C3b and BSA.

2.2. PEG chain length

PEG chain length is another important parameter that plays a crucial role in

both protein adsorption and complement activation. Mosqueira et al.21 dem-

onstrated that polymeric particles covered with a high density of 20-kDa PEG

chains show the lowest protein adsorption and lowest complement activation

among the materials they tested. This is presumably due to steric barriers sur-

rounding these particles. Mosqueira’s team hypothesized that the long PEG

chains act through creating a PEG-hydrated cloud “shielding” negatively

charged groups located underneath it. In addition, they reported that

particles coated with 5-kDa PEG with high density are more effective in

reducing complement activation than at low density.



Recent studies on PEG–PLGA nanoparticles further corroborate that

long chain lengths reduce protein adsorption and complement activation.22

Research on the in vivo effects of short and long chains of monolayer-

protected gold nanoparticles indicated that the former presented short

circulation half-life, whereas longer half-life was observed with long PEG

chains.23

2.3. Morphology of the nanocarrier

Activation of the complement system has been assessed on nanomaterials

that are either spherical or tubular, including liposomes, micelles, lipid-

polymer nanoparticles, and carbon nanotubes. Although these studies are

signs of progress on this topic, there still lacks a deep understanding of the

influence of the nanocarrier’s shape on complement activation. To the best

of our knowledge, the only insight on this topic comes from a study that

suggested that the higher the curvature, the lower the human plasma pro-

tein surface coverage, thus rendering activation of the complement system

less likely.24 For spherical nanocarriers, the curvature is directly related

to the size of the particles.24 Small nanoparticles (hydrodynamic diameter

of ~70 nm) have higher curvature than large nanoparticles (hydrodynamic

diameter of ~300 nm). Despite this finding, the morphology of the nano-

carrier itself seems to be irrelevant for the activation of the complement

system; rather, it is the surface chemistry of the nanocarrier that is the key

to complement activation. As a matter of fact, it has been demonstrated

that lipid PEGylated carbon nanotubes activate the complement system

regardless of the terminal functional group and the PEG brush-like surface

structure on HIPco SWNTs.25 The activation of the complement system by

lipid PEGylated HIPco SWNTs might be due to the incomplete coverage

of PEG on their surface. Since the un-PEGylated surface area of HIPco

SWNTs is hydrophobic, it is likely that C1q molecules bind to that surface

via hydrophobic interactions. Previous studies on non-chemically modified

HIPCo SWNTs showed the activation of the complement system via the

classical pathway, which is initiated by the binding of C1q to the activator

surface.3

In the field of polymeric nanoparticles, Gref et al.26 reported the influence

of the core composition on protein adsorption. Protein binding studies were

conducted with nanoparticles synthesized with three different polymeric

cores made of PLGA, PLA, and poly(ε-caprolactone) (PCL) polymers.

During this synthesis, the length and density of the PEG were kept constant.

The results showed slight differences in the pattern of protein binding which



suggested that the core has a part in determining the proteins that bind to

the nanoparticle surface.

3. Recent Engineering Approaches to Avoid or Reduce Complement Activation

Understanding the factors that activate the complement system is very impor-

tant to engineering biocompatible nanocarriers that can be used for a wide

range of applications in medicine. In the last two decades, there has been

great progress in understanding the complement system cascade and its inter-

action with different organic and inorganic materials. The complement sys-

tem views liposomes, carbon nanotubes, polymeric nanoparticles, and

micelles as invaders. Synthetic strategies such as the use of PEG to coat the

surface of these nanocarriers to evade the complement system cascade have

been employed.18,25,27 This chemical approach has worked only to a certain

extent for immune evasion. For instance, studies have demonstrated that

PEGylation does not necessarily suppress complement opsonization.27

Complement activation induces the opsonization of PEGylated liposomes

because of the covalent deposition of C3b on the liposomal surfaces.

In previous sections, we mentioned that PEG can present two different

spatial arrangements known as the mushroom- and brush-like configurations.

These configurations influence protein adsorption which might or might not

lead to complement activation. Lately, it has been found that the transition

from one configuration to another influences the activation of the comple-

ment system.28

The reduction or prevention of complement activation is relevant for

several reasons. Preventing the activation of the complement system can help

to extend the circulation half-life of the nanocarrier in the bloodstream,

improve the nanocarrier’s biodistribution profile, and avoid a set of allergies

known as CARPA (complement activation-related pseudoallergy).29 For all

these reasons, it is urgent to develop new synthetic strategies for diminishing

or preventing complement system activation. We next discuss the most recent

advances in this area.

3.1. Influence of PEG mushroom- and brush-like configurations on complement activation

Surface PEGylation not only helps diminish protein adsorption, but can also

cause alterations of the pathway for complement activation. Moghimi et al.28

reported that the transition of copolymer architecture on nanoparticles with



polyethylene oxide (PEO) chains from the mushroom–brush to brush-like

configuration not only switches complement activation from the

C1q-dependent classical to lectin pathway, but also reduces the level of gener-

ated complement activation products C4d, Bb, C5a, and SC5b-9.

According to the authors, changes in adsorbed polymer configuration

trigger alternative pathway activation differently and through different initia-

tors. This was the first study to demonstrate the importance of configura-

tional mobility of surface-projected PEG chains in the modulation of

complement activation with a spectrum of model nanoparticles that exhibited

different pharmacokinetic profiles.

The mushroom–brush and brush-like PEG configurations are observed

on the surface of different nanocarriers, including liposomes,29 polymeric

nanoparticles,26 gold nanoshells30 and carbon nanotubes.25 In all these cases,

the brush-like configuration has been identified as the surface structure that

reduces protein adsorption, though it does not necessarily diminish comple-

ment activation. PEGylated liposomes are a good example of this case.

Bradley et al.31 found that the incorporation of PE–mPEG [phosphatidyletha-

nolamine–monomethoxylpoly(ethylene glycol)2000] from 5–7.5 mol% into

the liposomal bilayer was not enough to prevent complement activation.

Their results revealed that the inhibitory effect of mPEG–lipid on comple-

ment activation is highly dependent on the liposomal concentration used in

the complement assay.31 Other studies have reported that the concentration

of PEG in liposomes causes the transition from the mushroom- to brush-like

regime.32 Liposomes prepared with up to 4 mol% of grafted PEG exhibited

the mushroom configuration because the neighboring coils did not interact

laterally. On the other hand, liposomes synthesized with PEG concentration

above 4 mol% showed a brush regime since the neighboring PEG chains

pushed against each other, extending farther out from the surface on which

they were grafted.32 This study did not report on complement system

activation under these circumstances.

The spatial configuration of PEG on polymeric nanoparticles has been stud-

ied for more than a decade. It is well known that PEG can exhibit either the

mushroom- or brush-like configuration on a polymeric nanoparticle surface.

PEG with high molecular weight and at high density is expected to exhibit the

brush-like configuration.33 Mosqueira et al.21 demonstrated that 20-kDa PEG

chains are more effective in preventing C3 cleavage than 5-kDa PEG chains

because of the steric barrier created by the PEG surrounding the particles.

Core-shell nanoparticles synthesized with di-block copolymer of

methoxyPEG–PCL (MePEG–PCL) and tri-block copolymer of PCL–PEG–

PCL present mushroom- and brush-like surface structures, respectively. The



mushroom-like structure was thought to be more efficient in preventing

opsonization because it could form a more effective conformational cloud.34

Recently, a quantitative and nondestructive assay based on surface enhanced

Raman scattering (SERS) spectroscopy has been reported to determine the

number of PEG molecules bound to gold nanoshell surfaces.30 From this

estimate, one can obtain the pack density of PEG units on nanoshell surfaces,

which helps to infer their configuration.30

Heterogeneity is a phenomenon that sooner or later will have impact on

complement activation, and therefore must be taken into account when

designing new nanocarriers. The concept of surface heterogeneity in essence

describes the incomplete coverage of the nanoparticle’s surface with PEG.35

Moghimi and Szebeni discussed the direct effect of the partial coverage of

PEG on the nanoparticles’ circulation half-life in the bloodstream.35 The less

the surface coverage, the poorer the steric shielding, and the shorter the cir-

culation life, which is presumably caused by the binding of opsonic molecules

to the unshielded area.4 The higher the surface coverage, the greater the

resistance to protein binding. For instance, microsphere populations covered

with a high density of mPEG with the mushroom–brush intermediate and/

or brush-like configurations were most resistant to phagocytosis and activated

the complement system poorly.36

To date, there has been little discussion of the influence that factors such

as temperature,37 autoxidation catalyzed by transient metals,38 and salt con-

centration39 might have on the configuration and stability of PEG. Changes

in the physicochemical characteristics of PEG might affect PEG configura-

tion. There is evidence that the aforementioned parameters have a direct

effect on PEG. As a matter of fact, it has been observed that increasing the

salt concentration to 3 M and raising the temperature to 37°C resulted in the

rapid aggregation of PEGylated nanoparticles.40 This is due to the fact that

high temperature causes the dissolution of the PEG hydration layer, leading

to particle precipitation. Although the assessment of complement activation

and protein binding studies were not conducted at such high salt concentra-

tions, these parameters should be kept in mind until new studies report the

temperature and salt concentration at which mushroom- and brush-like con-

figurations are not only formed, but are also stable.

The identification of these parameters is very important for the efficient

design of nanocarriers, especially when application is envisioned in the medi-

cal field. Depending on the application, the experimental conditions in which

the nanocarriers will be immersed might be subject to high temperature and

high salt concentration. Temperature is the one of the parameters that might

have a strong influence in complement activation depending on the method



used. For example, hemolytic assays are conducted at 37°C and a great num-

ber of papers has described the assessment of complement activation using

hemolytic assays (which can only be done between about 20°C and 37°C).

New synthetic approaches are already available as an alternative to PEG,

which is traditionally the workhorse for preventing protein adsorption.40

These novel polymers are not temperature-sensitive and the surface structure

does not change at high salt concentration.41 However, they are not as well

studied and characterized as PEG since they have emerged only recently.

3.2. Modulation of complement system activationvia functional groups

For more than 30 years, scientists have been trying to prevent or diminish

complement activation to achieve blood biocompatibility. In the 1980s and

1990s, there were reports of various methods to diminish complement activa-

tion.41–45 Carreno et al.45 reported that the substitution of hydroxyl groups of

Sephadex® (an activator of the alternative pathway) by carboxymethyl (CM),

groups can reduce the activating capacity of the resulting polymer (CMSeph).

Complement activation by CMSeph can be abolished when an average of one

CM group is present per glycosyl unit. However, this method is a multistep

procedure and involves complicated organic synthesis.

Another example of reducing activation of the complement system via

direct chemical modification of the nanomaterial is in the field of carbon

nanotubes. In previous studies, we reported that the chemical modification

of pristine MWNTs reduces complement activation.46 In this study, four dif-

ferent types of chemically modified MWNTs were tested for complement

activation via the classical and alternative pathways using hemolytic assays.

It was found that MWNTs functionalized with e-caprolactan or L-alanine

showed, respectively, >90% and >75% reduction in classical pathway activation

compared with unmodified MWNTs. The reduced level of complement acti-

vation via the classical pathway that is likely to increase biocompatibility is

directly correlated with the amount of C1q protein bound to chemically

modified carbon nanotubes. These results demonstrated for the first time that

these types of chemical modifications are able to considerably alter the level

of specific complement proteins bound by pristine MWNTs.

Lately, new synthetic methods have been developed not only to diminish

complement system activation, but most importantly, to manipulate it at will.

In this method, the manipulation of the complement system is due to the

modulation of the nanoparticles’ surface charge. For some nanocarriers, the

surface charge dictates the activated pathway. This is the case with liposomes.



Negatively charged liposomes containing phosphatidyl glycerol, phosphatidic

acid, cardiolipin, and/or phosphatidylinositol activate the complement sys-

tem via the classical pathway.47 Positively charged liposomes containing stear-

ylamine activate the alternative pathway.47 Neutral liposomes hardly activate

the complement system.47

For other potential drug delivery systems such as PEGgylated lipid carbon

nanotubes and polymeric nanoparticles, the role of surface charge is still rele-

vant for complement activation, but differences among liposomes and in

charges in carbon nanotubes and polymeric nanoparticles do not determine

which pathway will be activated. Results reported by Hamad et al.25 demon-

strated that PEGylated lipid HIPco SWNTs activate the whole complement

system independently of the terminal end moiety of the projected PEG chain.

The authors hypothesized that HIPco SWNTs most likely activate the comple-

ment system via the lectin pathway. In addition, PEGylated lipid polymeric

nanoparticles functionalized with NH2, COOH, and CH3 activated the alter-

native pathway, but not the classical pathway.48 This scientific evidence sug-

gested that the presence of functional groups in the PEG chain does not

necessarily determine the pathway of the complement system that is activated.

So far, scientists have not yet determined the set of parameters that abso-

lutely dictate the pathway activated. In an effort to shed light on this matter,

Toda et al.49 conducted studies assessing complement activation of PEGylated

self-assembly monolayers (SAMs) functionalized with a mixture of NH2–

COOH and NH2–CH3. Briefly, functionalized SAMs were prepared with vari-

ous molar ratios of a pair of NH2,–COOH or NH2–CH3 on gold-coated glass

plates. These coated glass plates were immersed in reaction mixtures for 24 h

followed by several washes. Their results showed that the NH2–COOH mix-

ture activated the complement system, whereas the NH2–CH3 pair did not

activate it. The authors believed that the NH2–CH3 mixture did not activate

the complement system because of the numerous serum proteins adsorbed

onto those SAMs, including albumin, that formed a protein layer which inhib-

ited access of C1q or C3b to the surface. On the contrary, a high amount of

C3b or C3 convertase was found to be deposited on the NH2–COOH SAM.

We showed a direct correlation between the zeta potential, levels of com-

plement activation, and the amount of C3b deposited on lipid–polymeric

nanoparticles functionalized with a mixture of NH2–COOH and NH2–CH3

and synthesized via the nanoprecipitation method.48 We observed that as the

presence of NH2 on PEGylated lipid–polymer nanoparticles reduces, the level

of complement activation diminishes and the zeta potential becomes more

negative. The C3b binds to NH2 and OCH3 as well as the different molar

ratio mixtures of these two functional groups. The molar ratio mixture of



COOH–NH2 shows complement activation, which agrees with Toda’s find-

ings, even when those studies were conducted on SAMs and not in polymeric

nanoparticles. This pair of functional groups mixtures also binds the C3 β

chain, which might explain the mechanism involved in the activation of the

complement system. The main difference between Toda’s findings and our

results is that we found that lipid–polymeric nanoparticles functionalized with

several molar ratio mixtures of NH2–CH3 activate the complement system.48

This could probably be due to the partial coverage of NH2–CH3 on the nano-

particle surface. Our findings clearly demonstrated a direct correlation of the

surface charge of the particles and the activation of the complement system.

In addition, our team’s method was a simple, practical, and inexpensive way

to modulate the complement system. Using this method, we can make these

particles act as adjuvants by functionalizing their polymeric core with DSPE–

PEG–NH2 or make them function as biocompatible nanocarriers for drug

delivery purposes by functionalizing their surfaces with CH3.

Hydroxyl groups are another terminating group that has been used to

functionalize SAMs to study their effects on complement activation. Sperling

and co-workers reported that these surfaces strongly activate the complement

system.50 As the amount of surface OH increases, the amount of C5a generated

also increases.50 Other studies also reported the activation of the complement

system induced by terminal hydroxyl group of tri(ethylene glycol)-terminated

alkanethiol (HS-TEGOH).51 Again, the cause of such activation was the depo-

sition of C3b on these surfaces.

4. Methods for the In Vitro Study of Complement Activation by Different Nanomaterials

4.1. Hemolytic assay (CH50)

For several decades, hemolytic assays have been used as the standard proce-

dure to assess complement activation. CH50, or total hemolytic complement

assay, measures the ability of the classical pathway-activated MAC to lyse

sheep red blood cells (SRBCs) coated with an antibody.52 The alternative

pathway hemolytic assay (APH50) measures the ability of the MAC generated

by this pathway to lyse rabbit red blood cells. Both assays indicate a deficiency

of a complement component by the absence of lysis. Hemolytic assays have

several advantages over other complement tests. For example, the biggest

advantage of the CH50 assay is the possibility of evaluating the activation of

the complement system in most mammalian species’ sera. Also, this method

is inexpensive. However, limitations include their labor intensity (particularly



the pipetting of small volumes), the short shelf-life of SRBCs, the variable

inter-lot performance of SRBCs, and the low sensitivity of the assay relative

to measuring C scission products such as C5a and SC5b-9.

Complement hemolytic assays can vary widely in sensitivity from day to

day, so it is possible to compare only results obtained in the same assay and

not results obtained on different days. This is mainly because of metabolic

changes in the cells, which make them more sensitive to lysis as they age. The

most common assay for complement consumption is done in a buffer con-

taining Ca2+ and Mg2+ ions. This measures classical pathway activation. If Ca2+

ions are removed by the chelator EGTA, the classical and lectin pathways are

completely inhibited, and only the alternative pathway is measured. There are

numerous versions of the hemolytic CH50 assay, which is arbitrarily standard-

ized with respect to the concentration of sensitized SBRCs (108/1.5 mL),

and the concentration and type of sensitizing antibody (heterophilic

Forssmann, i.e., rabbit anti-RBCs).

Hemolytic assays have also been used to evaluate the activation of the

complement system by different nanomaterials such as liposomes47 and pris-

tine and chemically modified carbon nanotubes.3,46 It is worth noting that

this simple complement consumption assay can only be used if the activator

is particulate and can be separated from the serum by filtration or centrifuga-

tion before complement activity is assayed in the serum. If the activator is

soluble and cannot be easily separated from the serum, more sophisticated

assays are required to distinguish between complement consumption and

inhibition of the complement assay.

4.2. ELISA kits

Lately, ELISA-like assays have been used to assess in vitro complement acti-

vation.25,48–49 In these studies, complement activation was assessed using

Quidel kit SC5b-9. This ELISA-based assay measures the cleavage of C5 and

subsequent terminal pathway activation. Specifically, SC5b-9 is generated by

the assembly of C5–C9 as a consequence of complement activation via all

three pathways and subsequent binding to the naturally occurring regulatory

serum protein, the S protein (vitronectin). SC5b-9 forms when the MAC

(Figure 1) fails to insert into a lipid bilayer, but instead reacts with S protein.

C9 within this complex expresses a neo-epitope (i.e., an epitope not present

in C9 which is not incorporated in the SC5b-9 complex), so C9 itself does

not interfere in the assay. ELISA-like assays have several advantages over

hemolytic assays. For instance, the former is a faster and easier method than

the latter. These methods do not involve the use of SRBCs so they eliminate



problems such as short shelf-life and inter-lot variability and sensitivity. For

these reasons, the ELISA-like assay method is superior to hemolytic assays.

In addition, this method can be adapted to assess complement activation in

a high-throughput screening fashion. On the other hand, disadvantages of

ELISA-based assays include the evaluation of the complement system only in

human sera and not in other mammalian species (although a few reagents,

Figure 3. Schematic diagram showing the main steps in hemolytic assays. This diagram

illustrates the assessment of the complement system via the classical pathway. This type of assay

measures the ability of the classical pathway-activated membrane attack complex (MAC) to lyse

sheep red blood cells (SRBCs) coated with an antibody. Likewise, the functional activity of the

alternative pathway and the terminal components can be measured by the lysis of rabbit eryth-

rocytes in human serum. (a) Assessing the activation of the complement system is done by

simply incubating the nanomaterial (e.g., carbon nanotubes) with serum. Samples are incubated

at 37°C for 1 h followed by centrifugation. (b) The supernatant of the sample is serially diluted

and placed in a microtiter plate. One hundred microliters of each dilution are incubated with

100 µL of antibody-sensitized SRBCs (EA) (108 cells/mL in veronal buffer). SRBCs sensitized

with an antibody (EA) are used as an immune complex to activate the classical pathway. When

EA cells are added to serum or plasma, they are lysed as a result of their activation of the comple-

ment system and the intercalation of C5b-9 MAC into their cell membranes. (c) After incuba-

tion, cells are spun down (2500 rpm, 10 min, room temperature) and hemoglobin is measured

at 405 nm in the supernatant. (d) Calculating complement activation is done by plotting the

percent lysis against the dilution of human serum (logarithmic scale). (Adapted from Ref. 52.)



e.g., antibodies, are available for mice or rats). Another disadvantage is that

this assay only indicates whether the complement system is activated or not.

Further experiments are needed to identify the activated complement system

pathway. For this purpose, there are similar kits that can be used. For

instance, Quidel C4d and Bb ELISA kits are specific for classical/lectin and

alternative pathway complement activators, respectively. Both analytes are

by-products of complement activation; C4d is a scission derivative of C4b,

whereas Bb rises in blood as a consequence of spontaneous dissociation of

the alternative pathway C3 convertase. Other assay kits which measure C3a

or C5a generation are available from Enzo Life Sciences, BD Biosciences,

R & D systems, and others. All of these kits are relatively expensive and they

expire in less than a year.

Figure 4. Schematic diagram of the Quidel-kit as an example of Elisa-like assays.

(a) Washing assay wells with wash solution as indicated in Quidel kit SC5b-9. (b) Pipetting

specimen diluents (black), standards, controls, and specimens into assay wells followed by

incubation and several washes. (c) Pipetting SC5b-9 with the conjugate into assay wells fol-

lowed by incubation and several washes. (d) Pipetting substrate into assay wells. (e) Adding

stop solution into assay wells. (f) Reading the optical density at 450 nm and analyzing the assay

results using a linear curve fit (y = mX + b). (Adapted from Ref. 48.)



4.3. Wieslab diagnostics kits

Wieslab diagnostics kits are also ELISA-like assays for the qualitative determi-

nation of functional classical, alternative, and lectin pathways in human

serum.53 Wieslab products are available via Eurodiagnostica (http://www.

eurodiagnostica.com/). These complement assays combine the principles of

the hemolytic assay for complement activation with the use of labeled anti-

bodies specific for the SC5b-9 neoepitope produced as a result of comple-

ment activation. The amount of SC5b-9 generated is proportional to the

functional activity of complement pathways. The wells of the microtiter strips

are coated with specific activators of the classical, lectin (MBL-only), or the

alternative pathway. Patient serum is diluted in diluents containing specific

reagents to ensure that only the appropriate pathway is activated. During the

incubation of the diluted patient serum in the wells, complement is activated

by the specific coating. It is worth noting that the level of complement activ-

ity evaluated by functional assays such as the Wieslab complement kits takes

into account the rate of synthesis, degradation, and consumption of the com-

ponents, and provides a measure of the integrity of the pathways as opposed

to immunochemical methods, which specifically measure the concentration of

various complement components.

4.4. 2D immunoelectrophoresis method

The 2D immunoelectrophoresis method is another technique that has been

used to assess complement system activation.12 This technique separates and

characterizes proteins based on electrophoresis and reaction with antibod-

ies. The cleavage of C3 into breakdown products C3b, iC3b, and C3c alters

the electrophoretic mobility of C3. C3 is separated from its breakdown

products by electrophoresis on agarose, and the proteins are electro-

phoresed at right angles into an agarose gel containing anti-C3 antibodies.

The proteins react with antibodies and form a visible precipitate over an

area proportional to the protein concentration. This is a time-consuming

procedure, unsuitable for high-throughput screening, but for a one-time

assessment, it could be a good option. Also, this technique is not as sensitive

for complement activation assessment as ELISA-like assays. However, this

technique could be less expensive than ELISA-Quidel Kits and hemolytic

assays as the reagents for the test can be used for other research purposes

and be frozen for future use. The long shelf-life of antibodies against

complement proteins is one of the biggest advantages of the 2D



immunoelectrophoresis method as both ELISA-like assays and CH50 have

short shelf-lives. For example, reagents for hemolytic assays such as SRBCs

only last for several days, whereas ELISA kits last for 12 months. Table 1

summarizes all the in vitro assays employed to assess complement activation

described above.

Figure 5. Schematic diagram of activation of the complement system evaluated from 2D

immunoelectrophoresis method. (a) Incubation of the nanomaterial (e.g., nanoparticles) with

human serum in veronal buffer supplemented with MgCl and CaCl. (b) 2D electrophoresis gel

loaded with incubated samples as described in Ref. 12. (c) Coomassie blue staining of 2D elec-

trophoresis gel. (d) Analysis of the immunoelectrophoregrams showing the presence of C3 and

C3b proteins after incubation of the serum with the nanomaterial. (Adapted from Ref. 12.)

Table 1. Summary of in vitro studies of complement activation on

different nanomaterials.

Assay method Assay source/protocol

Hemolytic assays (CH50) Salvador-Morales, et al.3

Gbadamosi J K. et al.4

Whaley K54

ELISA-like kits Quidel Corp., San Diego, CA

Enzyme immunoassays Ferraz N et al.55

Wieslab http://www.eurodiagnostica.com/

2D immunoelectrophoresis Vauthier C et al.12

Bertholon I et al.56



5. In Vivo Studies of Complement Activation by Different Nanomaterials

To date, in vivo studies on complement system activation are still limited.29,57–60

Szebeni and co-workers have made substantial contributions on this topic by

assessing complement activation in different animal models including pigs,

dogs, and rats.29 Their findings showed that these animals present a set of

allergy symptoms when they are injected with lipid-based nanomaterials. These

allergy symptoms include tachypenia, tachycardia, hypotension, hypertension,

chest pain, and back pain. These reactions, commonly known as infusion reac-

tions, form the concept of CARPA.60 Evidence from in vitro, in vivo, and clini-

cal trials suggest that these reactions are derived from complement system

activation and different mammalian species respond differently to the intrave-

nous (i.v.) administration of liposomes. For example, rats are less sensitive than

dogs and pigs; pigs are the most sensitive animal model for the i.v. injection of

lipid-based materials.60 Szebeni and co-workers’ findings also show that the

presence of cholesterol and phospholipids in liposomes are responsible for com-

plement system activation.61–62 In addition, it is well known that cationic and

anionic liposomes activate the complement system via the alternative and clas-

sical pathways, respectively.47 In this case, it is reported that the complement

system is activated not only due to the presence of cholesterol and phospholip-

ids, but also cationic and anionic molecules in liposomes.

Other lipid-based materials such as micelles cause infusion reactions in

dogs and pigs.60 However, other studies demonstrated that mPEG–

phosholipid in micelles and liposomes coated with methoxyl functional

groups did not activate the complement system. These results indicated that

charge is the property of the liposomes that causes complement activation.

PEGylated HIPco SWNTs with CH3 and NH2 terminal ends also induce in

vitro and in vivo complement activation.25 In this study, the in vivo comple-

ment activation assessment was measured indirectly by the levels of throm-

boxane B2 in rat blood.

The differences in complement system activation across mammals are

based on species, dose dependence, and the influence of lipid composition.

For instance, the mechanism involved in the activation of the complement

system in pigs that are injected with Doxil® is the anaphylaxis phenomenon,

which is basically the increment of C5a and C3a elements. This increment

causes severe cardiac abnormalities in porcine models. On the other hand, a

similar amount of undiluted Doxil® can be fatal for men and dogs. Rats are

two to three orders of magnitude less sensitive to liposomes, at least to those

containing <50% cholesterol,42 although complement-dependent shock and



tissue damage induced by the i.v. injection of cholesterol-enriched liposomes

in rats have been observed.

Lipid composition plays an important role in complement activation.

Szebeni and co-workers reported that the cholesterol content of liposomes

may be an important determinant of high sensitivity reactions (HSR).

Pulmonary hypertensive effects of multilayer vesicles in pigs were found to be

proportional to the amount of cholesterol in the vesicle in the 20–71% range,63

and rat murine leukemic virus (MLV) containing 71% cholesterol were signifi-

cantly more reactogenic compared to liposomes with 45% cholesterol.42

Recently, Szebeni and co-workers have reported the activation of the com-

plement system in porcine models because of the presence of polyethylenimine

polymers.64 This type of polymer triggers anaphylactic reactions in low percent-

ages of hypersensitive individuals regardless of PEGylation. The results should

be kept in mind when engineering new drug delivery systems, as transient and

mild reactions can be fatal for patients who suffer from allergies and heart dis-

eases.65 More studies are urgently needed to be conducted on this topic to fully

elucidate the mechanisms involved in in vivo complement activation.

6. Concluding Remarks

Complement system activation can be triggered by the direct binding of key

complement proteins such as C1q, MBL, ficolins, and C3b to different nano-

particle surfaces. The binding of other plasma proteins such as fibrinogen and

BSA can serve as a bridge for the binding of proteins such as C3b, which in

turn activates the complement system. Thus, protein adsorption undoubtedly

plays a significant role in complement activation. Therefore, there is a great

need to reduce protein adsorption by understanding the parameters of the

nanosurfaces that enhance or avoid protein binding.

In this chapter, we have highlighted the physicochemical characteristics of

both complement proteins and nanomaterials that cause complement activa-

tion. Parameters such as the PEG chain length, density, and conformation

strongly influence complement activation. The process of engineering nano-

carriers with a long circulation half-life and a good pharmacokinetic profile

involves making several trade-offs.

After more than three decades of research in this area, we still have not

produced a recipe that guarantees the synthesis of ideal nanocarriers. The

existing literature on this topic undoubtedly has helped us to much better

understand the parameters that we need to consider to engineer the ideal

nanocarrier. We have remarked that future decisions and trade-offs will

depend on the severity of the disease and the willingness of patients to take



the risk. As shown above, lipid-based nanocarriers activate the complement

system in vitro and in vivo. Most importantly, clinical trials have demon-

strated the activation of the complement system. These studies should not be

overlooked, but should be seriously considered when Doxil® and other lipid-

based drugs are consumed by patients who have heart diseases. Future devel-

opments in the area of biomaterials await to abrogate complement activation,

minimizing secondary effects. Bacteria have evolved ways of avoiding attack

by complement, e.g., by binding to their surface a complement downregula-

tory protein, factor h, from the host. It has recently been suggested that

mimicking bacteria such as by attaching factor h-binding peptides to nano-

material surfaces, may be a strategy worth exploring.66–67

References

1. Carroll MV, Sim RB. Complement in health and disease. Adv Drug Rev 2011;

63:965–975.

2. Muthusamy BHG, Suresh S, Rekha B, Srinivas D, Karthick L, Vrushabendra BM, Sharma

S, Mishra G, Chatterje P. Plasma proteome database as a resource for proteomics research.

Proteomics 2005;3531–3536.

3. Salvador-Morales C, Flahaut E, Sim E, Sloan J, Green ML, Sim RB. Complement

activation and protein adsorption by carbon nanotubes. Mol Immunol 2006;43:193–201.

4. Gaboriaud C, Juanhuix J, Gruez A, et al. The crystal structure of the globular head of

complement protein C1q provides a basis for its versatile recognition properties. J Biol

Chem 2003;278:46974–46982.

5. Arlaud GJ, Gaboriaud C, Thielens NM, Budayova-Spano M, Rossi V, Fontecilla-Camps JC.

Structural biology of the Cl complex of complement unveils the mechanisms of its activa-

tion and proteolytic activity. Mol Immunol 2002;39:383–394.

6. Kishore U, Gaboriaud C, Water P, et al. C1q and tumor necrosis factor superfamily:

Modularity and versatility. Trends Immunol 2004;25:551–561.

7. Ling WL, Biro A, Bally I, et al. Proteins of the innate immune system crystallize on carbon

nanotubes but are not activated. ACS Nano 2011; 5: 730–737.

8. Law SKA, Reid KBM. Complement. IRL Press, 1995.

9. Sim RB, Perkins SJ. Molecular modelling of C3 and its ligands. Curr Top Microbiol

Immunol 1990;153:209–222.

10. Sim RB, Sim E. Autolytic fragmentation of complement components C3 and C4 and its

relationship to covalent binding activity. Ann N Y Acad Sci 1983;421:259–276.

11. Malgorzata J, Rybak-Smith MJ, Pondman KM, Flahaut E, Salvador-Morales C, Sim RB.

Recognition of carbon nanotubes by the human innate immune system. In: Klingeler R,

Sim RB, eds. Carbon Nanotubes for Biomedical Applications. Springer-Verlag, 2011,

pp. 183–210.

12. Vauthier C, Persson B, Lindner P, Cabane B. Protein adsorption and complement activa-

tion for di-block copolymer nanoparticles. Biomaterials 2011;32:1646–1656.

13. Moghimi SM, Andersen AJ, Hashemi SH, et al. Complement activation cascade triggered

by PEG-PL engineered nanomedicines and carbon nanotubes: The challenges ahead. J

Controlled Release 2010;146:175–181.



14. Hed J, Johansson M, Lindroth M. Complement activation according to the alternative

pathway by glass and plastic surfaces and its role in neutrophile adhesion. Immunol Lett

1984;8:295–299.

15. Kazatchkine MD, Carreno MP. Activation of the complement system at the interface

between blood and artificial surfaces. Biomaterials 1988;9:30–35.

16. Chenoweth DE. Complement activation in extracorporeal circuits. Ann N Y Acad Sci

1987;516:306–313.

17. Toda M, Arima Y, Iwata H. Complement activation on degraded polyethylene glycol-

covered surface. Acta Biomater 2010;6:2642–2649.

18. Peracchia MT, Harnisch S, Pinto-Alphandary H, et al. Visualization of in vitro protein-

rejecting properties of PEGylated stealth (R) polycyanoacrylate nanoparticles. Biomaterials

1999;20:1269–1275.

19. Stolnik S, Illum L, Davis SS. Long circulation microparticle drug carriers. Adv Drug

Delivery Rev 1995;16:195–214.

20. Thomas SN, van der vlies AJ, O’ Neil CP, et al. Engineering complement activation on

polypropylene sulfide vaccine nanoparticles. Biomaterials 2011;32:2194–2203.

21. Mosqueira VC, Legrand P, Gulik A, et al. Relationship between complement activation,

cellular uptake and surface physicochemical aspects of novel PEG modified nanocapsules.

Biomaterials 2001;22:2967–2979.

22. Yang A, Liu W, Li Z, Jiang L, Xu H, Yang X. Influence of polyethyleneglycol modification

on phagocytic uptake of polymeric nanoparticles mediated by immunoglobulin G and

complement activation. J Nanosci Nanotechnol 2010;10:622–628.

23. Simpson CA, Agrawal AC, Balinski A, Harkness KM, Cliffel DE. Short-chain PEG mixed

monolayer protected gold clusters increase clearance and red blood cell counts. ACS

Nano 2011;5:3577–3584.

24. Cedervall T, Lynch I, Lindman S, et al. Understanding the nanoparticle–protein corona

using methods to quantify exchange rates and affinities of proteins for nanoparticles. Proc

Natl Acad Sci 2007;104:2050–2055.

25. Hamad I, Hunter C, Rutt KJ, Liu Z, Dai H. Complement activation by PEGylated single-

walled carbon nanotubes is independent of C1q and alternative pathway turnover.

Mol Immunol 2008;45:3797–3803.

26. Gref R, Luck M, Quellec P, et al. Stealth corona-core nanoparticles surface modified by

polyethylene glycol (PEG): Influences of the corona (PEG chain length and surface den-

sity) and of the core composition on phagocytic uptake and plasma protein adsorption.

Colloids Surf B Biointerfaces 2000;18:301–313.

27. Laverman P, Brouwers AH, Dams ET, Oyen WJ, Storm G, van Rooijen N. Preclinical and

clinical evidence for disappearance of long-circulating characteristics of polyethylene gly-

col liposomes at low lipid dose. J Pharmcol Exp Ther 2000;293:996–1001.

28. Hamad I, Al-Hanbali O, Hunter AC, Rutt KJ, Andresen TL, Moghimi SM. Distinct poly-

mer architecture mediates switching of complement activation pathways at the nano-

sphere-serum interface: Implications for stealth nanoparticle engineering. ACS Nano

2010;4:6629–6638.

29. Szebeni J, Alving CR, Rosivall L, et al. Animal models of complement-mediated hyper-

sensitivity reactions to liposomes and other lipid-based nanoparticles. J Liposome Res

2007;17:107–117.

30. Levin CS, Bishnoi SW, Grady NK, Halas NJ. Determining the conformation of thiolated

poly (ethylene glycol) on Au nanoshells by surface-enhanced raman scattering spectros-

copy assay. Anal Chem 2006;78:3277–3281.



31. Bradley AJ, Devine DV, Ansell AM, Janzen J, Brooks DE. Inhibition of liposome-induced

complement activation by incorporated poly(ethylene glycol)-lipids. Arch Biochem Biophys

1998;357:185–194.

32. Garbuzenko O, Barenholz Y, Priev A. Effect of grafted PEG on liposome size and on

compressibility and packing of lipid bilayer. Chem Phys Lipids 2005;135:117–129.

33. Buscall R. The stability of polymer latices. In: Buscall R, Corner T, Stagema JF, eds.

Polymer Colloids. Elsevier Applied Science, 1986, pp. 141–217.

34. Hu Y, Xie J, Tong YW, Wang CH. Effect of PEG conformation and particle size on the

cellular uptake efficiency of nanoparticles with the HepG2 cells. J Controlled Release

2007;118:7–17.

35. Moghimi SM, Szebeni J. Stealth liposomes and long circulating nanoparticles: Critical

issues in pharmacokinetics, opsonization and protein-binding properties. Prog Lipid Res

2003;42:463–478.

36. Gbadamosi JK, Hunter AC, Moghimi SM. PEGylation of microspheres generates a het-

erogeneous population of particles with differential surface characteristics and biological

performance. FEBS 2002;532:338–344.

37. Efremova NV, Sheth SR, Leckband DE. Protein-induced changes in poly(ethylene glycol)

brushes molecular weight and temperature dependence. Langmuir 2011;17:

7628–7636.

38. Luk YY, Kato M, Mrksich M. Langmuir. Self-assembled monolayer of alkane thiolates

presenting mannitol groups are inert to protein adsorption and cell attachment. Langmuir

2000;16:9604–9608.

39. Ataman M. Properties of aqueous salt solutions of poly(ethylene oxide). Cloud points, θ

temperatures. Colloid Polym Sci 1987;265:19–25.

40. Estephan ZG, Schlenoff PS, Schlenoff JB. Zwitteration as an alternative to PEGylation.

Langmuir 2011;27:6794–6800.

41. Crepon B, Maillet F, Kazatchkine MD, Jozefonvicz J. Molecular weight dependency of

the acquired anticomplementary and anticoagulant activities of specifically substituted

dextrans. Biomaterials 1987;8:248–253.

42. Baranyi L, Szebeni J, Savay S, et al. Complement-dependent shock and tissue damage

induced by intravenous injection of cholesterol-enriched liposome in rats. J Appl Res Clin

Exp Ther 2003,3:3.

43. Montdargent B, Maillet F, Carreno MP, Jozefowicz M, Kazatchkine M, Labarre D.

Regulation by sulphonate groups of complement activation induced by hydroxymethyl

groups on polystyrene surfaces. Biomaterials 1993;14:203–208.

44. Montdargent B, Toufik J, Carreno MP, Labarre D, Jozefowicz M. Complement activation

and adsorption of protein fragments by functionalized polymer surfaces in human serum.

Biomaterials 1992;13:571–576.

45. Carreno MP, Labarre D, Jozefowicz M, Kazatchkine MD. The ability of Sephadex to

activate human complement is suppressed in specifically substituted functional Sephadex

derivatives. Mol Immunol 1988;25:165–171.

46. Salvador-Morales C, Basiuk EV, Basiuk VA, Green ML, Sim R. Effect of covalent func-

tionalization on the biocompatibility characteristic of multi-walled carbon nanotubes.

J Nanosci Nanotechnol 2007;8:2347–2356.

47. Chonn A, Cullis PR, Devine DV. The role of surface charge in the activation of the

classical and alternative pathways of complement by liposomes. J Immunol

1991;146:4234–4241.



48. Salvador-Morales C, Zhang L, Langer R, Farokhzad OC. Immunocompatibility proper-

ties of lipid-polymer hybrid nanoparticles with heterogeneous surface functional groups.

Biomaterials 2009;30:2231–2240.

49. Toda M, Hirata I. Effects of hydrophobicity and electrostatic charge on complement

activation by amino groups. ACS Appl Mater Interfaces 2010;2:1107–1113.

50. Sperling C, Schweiss RB, Streller U, Werner C. In vitro hemocompatibility of self-assem-

bled monolayers displaying various functional groups. Biomaterials 2005;26:

6547–6557.

51. Arima Y, Toda M, Iwata M. Complement activation on surfaces modified with ethylene

glycol units. Biomaterials 2008;29:551–560.

52. Whaley K, North, J. Haemolytic assays for whole complement activity and individual

components. In: Dodds AW, Sim RB (eds.), Complement: A Practical Approach, IRL

Press, Oxford University Press, 1997, pp. 19–47.

53. Eelen MA, Roos A, Wieslander J, et al. Functional analysis of the classical, alternative, and

MBL Pathways of the complement system: Standardization and validation of a simple

ELISA. J Immunol Methods 2005;296:187–198.

54. Whaley K. Methods in complement for clinical immunologists. In: Whaley, D (ed.),

Measurement of Complement, Churchill Livingstone, 1985, pp. 77–140.

55. Ferraz N, Karlsson-Ott M, Hong J. Time sequence of blood activation by nanoporous

alumina: studies on platelets and complement activation. Microsc Res Tech 2010;73:

1101–1109.

56. Bertholon I, Vauthier C, Labarre C. Complement activation by core-shell poly

(isobutylcyanoacrylate)-polysaccharide nanoparticles: Influences of surface morphology,

length, and type of polysaccharide. Pharm Res 2006;23:1313–1323.

57 Szebeni J, Wassef NM, Spielberg H, Rudolph AS, Alving CR. Complement activation in

rats by liposomes and liposome-encapsulated hemoglobin: Evidence for anti-lipid and

antibodies and alternative pathway activation. Biochem Biophys Res Commun

1994;205:255–263.

58. Uetrecht J. Role of animal models in the study of drug-induced hypersensitivity reactions.

AAPS J 2006;7:E914-E921.

59. Szebeni J, Fontana JL, Wassef NM, et al. Hemodynamic changes induced by liposomes

and liposome- encapsulated hemoglobin in pigs: A model for pseudoallergic cardiopulmo-

nary reactions to liposomes: Role of complement and inhibition by soluble CR1 and anti-

C5a antibody. Circulation 1999;99:2302–2309.

60. Szebeni J. Complement activation-related pseudoallergy caused by amphiphilic drug car-

riers: the role of lipoproteins. Curr Drug Delivery 2005;2:443–449.

61. Moghimi MM, Hamad I, Bunger R, et al. Activation of the human complement system

by cholesterol-rich and PEGylated liposomes-modulation of cholesterol-rich liposome-

mediated complement activation by elevated serum LDL and HDL levels. J Liposome Res

2006;16:167–174.

62. Szebeni J, Wassef NM, Spielberg H, Rudolph AS, Alving CR. Complement activation in

rats by liposomes and liposome-encapsulated hemoglobulin: Evidence for anti-lipid anti-

bodies and alternative pathway activation. Biochem Biophys Res Commun

1994;205:255–263.

63. Szebeni J, Baranyi L, Savay S, et al. Complement activation-related cardiac anaphylaxis in

pigs: Role of C5a anaphylatoxin and adenosine in liposome-induced abnormalities in ECG

and heart function. Am J Physiol 2006;290:H1050-H1058.



64. Merkel OM, Urbanics R, Bedocs P, et al. In vitro and in vivo complement activation

and related anaphylactic effects associated with polyethylenimine and polyethylenimine-

graft-poly(ethylene glycol) block copolymers. Biomaterials 2011;32:4936–4942.

65. Szebeni J. Complement activation-related pseudoallergy: A new class of drug-induced

immune toxicity. Toxicology 2005;216:106–121.

66. Sim RB, Willis R. Surface properties: Immune attack on nanoparticles. Nat Nanotechnol

2011;6:80–81.

67. Wu YQ, Qu H, Sfyroera G, et al. Protection of nonself surfaces from complement attack

by factor H-binding peptides: Implications for therapeutic medicine. J Immunol

2011;186:4269–4277.

complement activation - nanomedicinelabofnano.gmu.edu/.../book-chapter-on-complement-system.pdf ·...

Documents